# Semantic Segmentation on Sensat Urban

Source: [Semantic Segmentation on Sensat Urban](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:54:01.175Z

Evaluation of Semantic Segmentation performance on the Sensat Urban dataset, reporting metrics including Overall Accuracy (OA) and mean Intersection over Union (mIoU).

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Semantic Segmentation |
| Dataset (including any stated split) | Sensat Urban |
| Metric | OA |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [OA](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=OA>): Higher is better; unit: PERCENT.

- [mIoU](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=mIoU>): Higher is better; unit: PERCENT.

- [Accuracy: Ground](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Ground>): Higher is better; unit: PERCENT.

- [Accuracy: Vegetation](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Vegetation>): Higher is better; unit: PERCENT.

- [Accuracy: Building](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Building>): Higher is better; unit: PERCENT.

- [Accuracy: Wall](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Wall>): Higher is better; unit: PERCENT.

- [Accuracy: Bridge](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Bridge>): Higher is better; unit: PERCENT.

- [Accuracy: Parking](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Parking>): Higher is better; unit: PERCENT.

- [Accuracy: Rail](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Rail>): Higher is better; unit: PERCENT.

- [Accuracy: Traffic](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Traffic>): Higher is better; unit: PERCENT.

- [Accuracy: Street](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Street>): Higher is better; unit: PERCENT.

- [Accuracy: Car](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Car>): Higher is better; unit: PERCENT.

- [Accuracy: Footpath](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Footpath>): Higher is better; unit: PERCENT.

- [Accuracy: Bike](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Bike>): Higher is better; unit: PERCENT.

- [Accuracy: Water](<https://www.sota2.com/research/sota/semantic-segmentation-on-sensat-urban.md?metric=Accuracy%3A%20Water>): Higher is better; unit: PERCENT.

## Recorded results

| Method | OA | Result date | Report paper |
| --- | --- | --- | --- |
| EyeNet | 93.7 | Not recorded | 68fa6fbe1e209131cea47309 |
| KPConv | 93.2 | Not recorded | 68fa6fbe1e209131cea47309 |
| RandLA++ | 91.9 | Not recorded | 68fa6fbe1e209131cea47309 |
| RandLA | 89.78 | Not recorded | 68fa6fbe1e209131cea47309 |
| SparseConv | 88.66 | Not recorded | 68fa6fbe1e209131cea47309 |
| SPGraph | 85.27 | Not recorded | 68fa6fbe1e209131cea47309 |
| PointNet++ | 84.3 | Not recorded | 68fa6fbe1e209131cea47309 |
| PointNet | 80.78 | Not recorded | 68fa6fbe1e209131cea47309 |
| Tangent Conv | 76.97 | Not recorded | 68fa6fbe1e209131cea47309 |

### EyeNet

OA: 93.7. Result ID: 6990ea4b3a6c31484904426b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### KPConv

OA: 93.2. Result ID: 6990ea4b3a6c314849044268.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66341e209131cea0ebe7.md>)

[Implementation](<https://github.com/HuguesTHOMAS/KPConv>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### RandLA++

OA: 91.9. Result ID: 6990ea4b3a6c31484904426a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### RandLA

OA: 89.78. Result ID: 6990ea4b3a6c314849044269.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67771e209131cea16862.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SparseConv

OA: 88.66. Result ID: 6990ea4b3a6c314849044267.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63d81e209131cea0140a.md>)

[Implementation](<https://github.com/facebookresearch/SparseConvNet>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SPGraph

OA: 85.27. Result ID: 6990ea4b3a6c314849044266.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642f1e209131cea03453.md>)

[Implementation](<https://github.com/loicland/superpoint_graph>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PointNet++

OA: 84.3. Result ID: 6990ea4b3a6c314849044264.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63641e209131ce9fe5a5.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PointNet

OA: 80.78. Result ID: 6990ea4b3a6c314849044263.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634f1e209131ce9fe019.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Tangent Conv

OA: 76.97. Result ID: 6990ea4b3a6c314849044265.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbe1e209131cea47309.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64851e209131cea051e1.md>)

[Implementation](<https://github.com/tatarchm/tangent_conv>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
